Poor implementation explains significant reductions in effect sizes when Evidence-Based Programs (EBPs) are translated to real-world settings (Henggeler, 2004). Multiple dimensions of implementation influence outcomes (Durlak & DuPre, 2008), but only four occur within the delivery of program sessions and, as a result, are potential modifiable sources of disconnect between the program as designed and as implemented: 1) fidelity (amount of curriculum providers deliver), 2) positive adaptations (the quality of unscriptd additions providers make during delivery), 3) quality of delivery (skill with which providers delivr material and interact with participants, and 4) participant responsiveness (level of program participation, such as participant attendance and program skill use). Identifying efficient methods to monitor these 4 dimensions (Schoenwald, et al., 2011) as well as understanding how they interrelate to influence outcomes (Berkel, Mauricio, et al., 2011) and change across sessions has important implications for maintaining EBP effects in real-world settings. The proposed research uses data from the NIDA-funded effectiveness trial of the New Beginnings Program (NBP; R01DA026874) and independent observer data that we will collect. NBP is a parenting intervention that improves youth substance use and mental health outcomes in response to parental divorce (Wolchik, et al., 2002) through effects on intervention-targeted parenting mediators (McClain, et al., 2010). Community providers will deliver the NBP to approximately 550 parents across 75 intervention groups through a partnership with 5 county-level courts in Arizona. The effectiveness trial will only collect independent observer data on 10% of sessions and only examine simple, direct effects of implementation on outcomes. This study builds on the NBP trial by collecting independent observer data for 100% of sessions, providing the required data to address 3 research aims that will inform critical questions in implementation measurement and theory. In Aim 1, we compare the reliability and predictive validity of fidelity, positive adaptation, and quality across independent observer ratings, provide self-report, and supervisor ratings. We also examine sampling strategies to identify the optimal quantity of data (e.g., 10%, 20%) that must be sampled to reliably assess fidelity, positive adaptations, and quality and if this is contingent on program activity type (e.g., didactic versus skills practice) or provider characteristics (e.g., experience, attitudes about EBPs). In Aim 2, we test theoretical frameworks explaining implementation effects on outcomes and examine the role of parent gender and ethnicity within these frameworks. Aim 2 builds on extant studies by testing: a) moderational and meditational hypotheses about the effects of provider behaviors (i.e., fidelity, adaptation, quality) on responsiveness and outcomes, b) the temporal precedence of provider behavior effects on responsiveness and outcomes, and c) the moderating effects of parent gender and ethnicity on model pathways. In Aim 3, we examine changes in fidelity, adaptation, and quality over time and the influence of provider characteristics on these changes.